Add a more scalable ETS ordered_set implementation

The current ETS ordered_set implementation can quickly become a
scalability bottleneck on multicore machines when an application updates
an ordered_set table from concurrent processes [1][2]. The current
implementation is based on an AVL tree protected from concurrent writes
by a single readers-writer lock. Furthermore, the current implementation
has an optimization, called the stack optimization [3], that can improve
the performance when only a single process accesses a table but can
cause bad scalability even in read-only scenarios. It is possible to
pass the option {write_concurrency, true} to ets:new/2 when creating an
ETS table of type ordered_set but this option has no effect for tables
of type ordered_set without this commit. The new ETS ordered_set
implementation, added by this commit, is only activated when one passes
the options ordered_set and {write_concurrency, true} to the ets:new/2
function. Thus, the previous ordered_set implementation (from here on
called the default implementation) can still be used in applications
that do not benefit from the new implementation. The benchmark results
on the following web page show that the new implementation is many times
faster than the old implementation in some scenarios and that the old
implementation is still better than the new implementation in some
scenarios.

http://winsh.me/ets_catree_benchmark/ets_ca_tree_benchmark_results.html

The new implementation is expected to scale better than the default
implementation when concurrent processes use the following ETS
operations to operate on a table:

delete/2, delete_object/2, first/1, insert/2 (single object),
insert_new/2 (single object), lookup/2, lookup_element/2, member/2,
next/2, take/2 and update_element/3 (single object).

Currently, the new implementation does not have scalable support for the
other operations (e.g., select/2). However, when these operations are
used infrequently, the new implantation may still scale better than the
default implementation as the benchmark results at the URL above shows.

Description of the New Implementation
----------------------------------

The new implementation is based on a data structure which is called the
contention adapting search tree (CA tree for short). The following
publication contains a detailed description of the CA tree:

A Contention Adapting Approach to Concurrent Ordered Sets
Journal of Parallel and Distributed Computing, 2018
Kjell Winblad and Konstantinos Sagonas
https://doi.org/10.1016/j.jpdc.2017.11.007
http://www.it.uu.se/research/group/languages/software/ca_tree/catree_proofs.pdf

A discussion of how the CA tree can be used as an ETS back-end can be
found in another publication [1]. The CA tree is a data structure that
dynamically changes its synchronization granularity based on detected
contention. Internally, the CA tree uses instances of a sequential data
structure to store items. The CA tree implementation contained in this
commit uses the same AVL tree implementation as is used for the default
ordered set implementation. This AVL tree implementation is reused so
that much of the existing code to implement the ETS operations can be
reused.

Tests
-----

The ETS tests in `lib/stdlib/test/ets_SUITE.erl` have been extended to
also test the new ordered_set implementation. The function
ets_SUITE:throughput_benchmark/0 has also been added to this file. This
function can be used to measure and compare the performance of the
different ETS table types and options. This function writes benchmark
data to standard output that can be visualized by the HTML page
`lib/stdlib/test/ets_SUITE_data/visualize_throughput.html`.

[1]
More Scalable Ordered Set for ETS Using Adaptation.
In Thirteenth ACM SIGPLAN workshop on Erlang (2014).
Kjell Winblad and Konstantinos Sagonas.
https://doi.org/10.1145/2633448.2633455
http://www.it.uu.se/research/group/languages/software/ca_tree/erlang_paper.pdf

[2]
On the Scalability of the Erlang Term Storage
In Twelfth ACM SIGPLAN workshop on Erlang (2013)
Kjell Winblad, David Klaftenegger and Konstantinos Sagonas
https://doi.org/10.1145/2505305.2505308
http://winsh.me/papers/erlang_workshop_2013.pdf

[3]
The stack optimization works by keeping one preallocated stack instance
in every ordered_set table. This stack is updated so that it contains
the search path in some read operations (e.g., ets:next/2). This makes
it possible for a subsequent ets:next/2 to avoid traversing some nodes
in some cases. Unfortunately, the preallocated stack needs to be flagged
so that it is not updated concurrently by several threads which cause
bad scalability.

1 files changed, 33 insertions, 7 deletions

diff --git a/erts/emulator/beam/erl_db.c b/erts/emulator/beam/erl_db.c
index c009a3bde8..1df972f4b6 100644
--- a/erts/emulator/beam/erl_db.c
+++ b/erts/emulator/beam/erl_db.c
@@ -90,7 +90,8 @@ enum DbIterSafety {
     ITER_SAFE         /* No need to fixate at all */
 };
 #  define ITERATION_SAFETY(Proc,Tab) \
-    ((IS_TREE_TABLE((Tab)->common.status) || ONLY_WRITER(Proc,Tab)) ? ITER_SAFE \
+    ((IS_TREE_TABLE((Tab)->common.status) || IS_CATREE_TABLE((Tab)->common.status) \
+      || ONLY_WRITER(Proc,Tab)) ? ITER_SAFE                             \
      : (((Tab)->common.status & DB_FINE_LOCKED) ? ITER_UNSAFE : ITER_SAFE_LOCKED))
 
 #define DID_TRAP(P,Ret) (!is_value(Ret) && ((P)->freason == TRAP))
@@ -359,6 +360,7 @@ typedef enum {
 
 extern DbTableMethod db_hash;
 extern DbTableMethod db_tree;
+extern DbTableMethod db_catree;
 
 int user_requested_db_max_tabs;
 int erts_ets_realloc_always_moves;
@@ -414,6 +416,15 @@ free_dbtable(void *vtb)
 			 tb->common.fixations);
 	}
 #endif
+    if (erts_atomic_read_nob(&tb->common.memory_size) > sizeof(DbTable)) {
+        /* The CA tree implementation use delayed freeing and the  DbTable needs to
+           be freed after all other memory blocks that are allocated by the table. */
+        erts_schedule_thr_prgr_later_cleanup_op(free_dbtable,
+                                                (void *) tb,
+                                                &tb->release.data,
+                                                sizeof(DbTable));
+        return;
+    }
 	erts_rwmtx_destroy(&tb->common.rwlock);
 	erts_mtx_destroy(&tb->common.fixlock);
 	ASSERT(is_immed(tb->common.heir_data));
@@ -1076,7 +1087,7 @@ BIF_RETTYPE ets_update_element_3(BIF_ALIST_3)
     DB_BIF_GET_TABLE(tb, DB_WRITE, LCK_WRITE_REC, BIF_ets_update_element_3);
 
     UseTmpHeap(2,BIF_P);
-    if (!(tb->common.status & (DB_SET | DB_ORDERED_SET))) {
+    if (!(tb->common.status & (DB_SET | DB_ORDERED_SET | DB_CA_ORDERED_SET))) {
 	goto bail_out;
     }
     if (is_tuple(BIF_ARG_3)) {
@@ -1165,7 +1176,7 @@ do_update_counter(Process *p, DbTable* tb,
 
     UseTmpHeap(5, p);
 
-    if (!(tb->common.status & (DB_SET | DB_ORDERED_SET))) {
+    if (!(tb->common.status & (DB_SET | DB_ORDERED_SET | DB_CA_ORDERED_SET))) {
 	goto bail_out;
     }
     if (is_integer(arg3)) { /* Incr */
@@ -1647,15 +1658,15 @@ BIF_RETTYPE ets_new_2(BIF_ALIST_2)
 	val = CAR(list_val(list));
 	if (val == am_bag) {
 	    status |= DB_BAG;
-	    status &= ~(DB_SET | DB_DUPLICATE_BAG | DB_ORDERED_SET);
+	    status &= ~(DB_SET | DB_DUPLICATE_BAG | DB_ORDERED_SET | DB_CA_ORDERED_SET);
 	}
 	else if (val == am_duplicate_bag) {
 	    status |= DB_DUPLICATE_BAG;
-	    status &= ~(DB_SET | DB_BAG | DB_ORDERED_SET);
+	    status &= ~(DB_SET | DB_BAG | DB_ORDERED_SET | DB_CA_ORDERED_SET);
 	}
 	else if (val == am_ordered_set) {
 	    status |= DB_ORDERED_SET;
-	    status &= ~(DB_SET | DB_BAG | DB_DUPLICATE_BAG);
+	    status &= ~(DB_SET | DB_BAG | DB_DUPLICATE_BAG | DB_CA_ORDERED_SET);
 	}
 	else if (is_tuple(val)) {
 	    Eterm *tp = tuple_val(val);
@@ -1716,7 +1727,13 @@ BIF_RETTYPE ets_new_2(BIF_ALIST_2)
     if (is_not_nil(list)) { /* bad opt or not a well formed list */
 	BIF_ERROR(BIF_P, BADARG);
     }
-    if (IS_HASH_TABLE(status)) {
+    if (IS_TREE_TABLE(status) && is_fine_locked && !(status & DB_PRIVATE)) {
+        meth = &db_catree;
+        status |= DB_CA_ORDERED_SET;
+        status &= ~(DB_SET | DB_BAG | DB_DUPLICATE_BAG | DB_ORDERED_SET);
+        status |= DB_FINE_LOCKED;
+    }
+    else if (IS_HASH_TABLE(status)) {
 	meth = &db_hash;
 	if (is_fine_locked && !(status & DB_PRIVATE)) {
 	    status |= DB_FINE_LOCKED;
@@ -3506,6 +3523,7 @@ void init_db(ErtsDbSpinCount db_spin_count)
 
     db_initialize_hash();
     db_initialize_tree();
+    db_initialize_catree();
 
     /* Non visual BIF to trap to. */
     erts_init_trap_export(&ets_select_delete_continue_exp,
@@ -4114,6 +4132,8 @@ static Eterm table_info(Process* p, DbTable* tb, Eterm What)
 	    ret = am_duplicate_bag;
 	} else if (tb->common.status & DB_ORDERED_SET) {
 	    ret = am_ordered_set;
+	} else if (tb->common.status & DB_CA_ORDERED_SET) {
+	    ret = am_ordered_set;
 	} else { /*TT*/
 	    ASSERT(tb->common.status & DB_BAG);
 	    ret = am_bag;
@@ -4409,6 +4429,12 @@ void erts_lcnt_enable_db_lock_count(DbTable *tb, int enable) {
 
     if(IS_HASH_TABLE(tb->common.status)) {
         erts_lcnt_enable_db_hash_lock_count(&tb->hash, enable);
+    } else if(IS_CATREE_TABLE(tb->common.status)) {
+        /* erts_lcnt_enable_db_catree_lock_count is not thread safe so
+           the table needs to get locked */
+        db_lock(tb, LCK_WRITE);
+        erts_lcnt_enable_db_catree_lock_count(&tb->catree, enable);
+        db_unlock(tb, LCK_WRITE);
     }
 }